典型文献
A Weighted Average Consensus Approach for Decentralized Federated Learning
文献摘要：
Federated learning(FedL)is a machine learning(ML)technique utilized to train deep neural networks(DeepNNs)in a dis-tributed way without the need to share data among the federated training clients.FedL was proposed for edge computing and Internet of things(IoT)tasks in which a centralized server was responsible for coordinating and governing the training process.To remove the design limitation implied by the centralized entity,this work proposes two different solutions to decentralize existing FedL algorithms,enabling the application of FedL on networks with arbitrary communication topologies,and thus extending the domain of application of FedL to more complex scenarios and new tasks.Of the two proposed algorithms,one,called FedLCon,is developed based on results from discrete-time weighted average consensus theory and is able to reconstruct the performances of the standard centralized FedL solu-tions,as also shown by the reported validation tests.
文献关键词：
中图分类号：
作者姓名：
Alessandro Giuseppi;Sabato Manfredi;Antonio Pietrabissa
作者机构：
Department of Computer,Control,and Management Engineering,University of Rome La Sapienza,Rome 00185,Italy;Department of Electrical Engineering and Information Technology,University of Naples Federico Ⅱ,Naples 80125,Italy
文献出处：
引用格式：
[1]Alessandro Giuseppi;Sabato Manfredi;Antonio Pietrabissa-.A Weighted Average Consensus Approach for Decentralized Federated Learning)[J].机器智能研究（英文）,2022(04):319-330
A类：
FedL,DeepNNs,decentralize,FedLCon
B类：
Weighted,Average,Consensus,Approach,Decentralized,Federated,Learning,learning,machine,ML,technique,utilized,deep,neural,networks,tributed,way,without,need,share,data,among,federated,training,clients,was,proposed,edge,computing,Internet,things,IoT,tasks,which,server,responsible,coordinating,governing,process,To,remove,design,limitation,implied,by,entity,this,proposes,different,solutions,existing,algorithms,enabling,application,arbitrary,communication,topologies,thus,extending,domain,more,complex,scenarios,new,Of,one,called,developed,results,from,discrete,weighted,average,consensus,theory,able,reconstruct,performances,standard,also,shown,reported,validation,tests
AB值：
0.584907
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
